Time to go negative

All right, I'm going to start trying to post regularly again. See if anyone's still reading.

Today's NY Times reports that "Stimulus by Fed is Disappointing, Economists Say." Let me quickly point out two problems with the article. First, the title. It's not really the passive voice, but it's in the classic circumlocutory style of the news media that I find annoying. I'd go with "Fed's Stimulus Disappoints Economists." Saves on newsprint too. Second, the article is very thinly sourced. On the Keynesian end of the spectrum we have Mark Thoma, on the RBC end we have Charles Plosser, and that's pretty much it. Add to that two finance professors from Northwestern and an economist from Bank of America. Not exactly a summary of the profession's opinion. And the article leaves the crucial question unanswered: does the fact that the Fed's efforts have not been tremendously successful mean that it should do more or less in the months to come? One could write a very interesting article surveying economists on their response to that question.

Put me in the "more" camp. Another round of fiscal stimulus, this time focused on aid to states to prevent them from continuing slashing budgets, would be the best policy. That's obviously not going to fly politically in Washington (nor, it has occurred to me these last few months, in the states - it now seems clear that Republican governors have found the fiscal crisis to be a very convenient pretext for pursuing their political objectives, beginning with smashing public employee unions). The Fed should certainly not end its bond purchasing program in June as scheduled, but it will face considerable political pressure to do just that, so I don't see much hope of any further quantitative easing.

That leaves one other policy response: imposing a negative interest rate on bank reserves. A number of economists have proposed this before. The idea is to penalize banks who hold excess reserves, force them to use those reserves to make loans or buy securities (which would have the effect of lowering interest rates, exactly like quantitative easing would do). Banks would also try to pass costs onto their customers by charging interest on checking deposits, so people with money in banks would have the incentive to spend rather than accumulate bank balances. The problem with this proposal is that faced with negative interest rates on checking deposits, customers might instead simply withdraw money from the banking system and hoard it in big piles on their dressers. After all, cash pays a zero percent interest rate which would be better than you could get at the bank.

Greg Mankiw half-seriously proposed imposing a negative interest rate on cash by once a year drawing a number between 0 and 9 by lot, and all currency with a serial number ending in that digit would no longer be useable as legal tender. This would in effect impose an expected rate of return of negative ten percent on all money holdings. (Incidentally: Greg Mankiw is a wonderful guy, a gracious host, a riveting speaker. But I find it highly ironic that this co-founder of the branch of macroeconomics known as New Keynesian economics had to have Silvio Gesell's proposal for stamped money, which Keynes cites approvingly in the General Theory, brought to his attention by Alan Taylor. Apparently Mankiw has not read the General Theory!)

Any scheme of stamped money or money-extinguishing lotteries is likely to be difficult to impose in practice. But there's an alternative. Hoarding money in response to negative interest rates on checking deposits is only possible if the Fed supplies currency perfectly elastically. This happens to be the Fed's current policy: if customers go to their banks in large numbers to withdraw money in cash from their checking accounts, the banks ask the Fed for more currency, and the Fed gives them as much as they want, exchanging the banks' reserves at the Fed for currency at par. But suppose the Fed simply exchanged reserves for currency at a discount (I suppose equal to the negative interest rate charged on reserves)? Banks would pass this charge on to customers: if you wanted to withdraw $100 of cash from your checking account where it was earning -5% interest, you'd have to pay a $5 fee for the privilege. The only way to avoid the charge would be to spend the money in your checking account in some way: buy groceries, a new car, a fine Easter bonnet; or if not that, put your funds into a money market account or retirement account where it ends up being used to purchase stocks, bonds, commercial paper, and other assets, thus driving the prices of those assets up and their yields down.

This proposal has the virtue of not requiring the Fed to purchase any more assets. In essence, the Fed forces the banking system and bank customers to engage in quantitative easing on its behalf. There may of course be practical difficulties. For one thing, the Fed would have to implement this plan (at least the fee for currency purchases) by surprise, because otherwise there would be a massive withdrawal of cash from the banking system in anticipation of the new fees. But the Fed employs thousands of very clever economists, I'm sure they could find their way around these problems.